ML20087P540
| ML20087P540 | |
| Person / Time | |
|---|---|
| Issue date: | 03/31/1984 |
| From: | Serkiz A Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| REF-GTECI-A-01, REF-GTECI-PI, TASK-A-01, TASK-A-1, TASK-OR NUREG-0993, NUREG-0993-R01, NUREG-993, NUREG-993-R1, NUDOCS 8404090143 | |
| Download: ML20087P540 (30) | |
Text
.-
NUREG-0993 Revision 1 Regulatory Analysis for USI A-1, " Water Hammer" (Formerly Value-Impact Analysis for USI A-1, " Water Hammer")
l I U.S. Nuclear Regulatory Commission Offica of Nuclear Reactor Regulation A. W. Serkiz
$ gbR RIGog'h
,g E
/
fy4 g j3 840331 0993 R pop
i
)
NOTICE Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following sources:
- 1. The NRC Public Docurnent Room,1717 H Street, N.W.
Washington, DC 20555
- 2. The NRC/GPO Sales Program, U.S. Nuclear Regulatory Commission, l
Washington, DC 20555
- 3. The National Technical Information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publications, it is not intended to be exhaustive.
l Referenced documents available for inspection and copying for a fee from the NRC Public Docu-ment Room include NRC correspondence and internal NRC memoranda; NRC Office of Inspection and Enforcement bulletins, circulars, information notices, inspection and investigation notices; Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and corredpondence.
i The following documents in the NUREG series are available for purchase from the NRC/GPO Sales Program: formal NRC staff and contractor reports, NRC-sponsored conference proceedings, and N RC booklets and brochures. Also available are Regulatory Guides, NRC regulations in the Code of Federal Regulations, and Nuclear Regulatory Comminion issuances.
Documents available from the National Technical Information Service include NUREG series esports and technical reports prepa ed by other federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.
Documents available from public and special technical libraries include all open literature items, such as books, journal and periodical articles, and transactions. Federal Register notices, federal and state legislation, and congressional reports can usually be obtained from these libraries.
l Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference i
proceedings are available for purchase from the organization sponsoring the publication cited.
Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the Division of Technical Information and Document Control, U.S. Nuclear Regulatory Com-mission, Washington, DC 20555.
Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bethesda, Maryland, and are available there for reference use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards Institute,1430 Broadway, New York, NY 10018.
GPO Printed copy pricc: -.13.75
NUREG-0993 Revision 1 Regulatory Analysis for USI A-1, " Water Hammer" (Formerly Value-impact Analysis for USI A-1, " Water Hammer")
ats u shed P a ch 984 A. W. Serkiz Division of Safety Technology Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Wcch!ngton, D.C. 20555
,/ %
I
l l
l l
ABSTRACT NUREG-0993, Revision 1 is the staff's regulatory analysis dealing with the resolution of the Unresolved Safety Issue A-1, Water Hamer.
This report contains the value-impact analysis for this issue, public comments received, and staff response, or action taken, in response to those comments. The staff's technical findings regarding water hammer in nuclear power plants are contained in NUREG-0927.
i I
l I
q h
l l'
4 4
i iii
CONTENTS PAGE Abstract lii I.
The Recommended Actions A.
Summary of Problem and Recommended Actions 1
B.
Need for the Recommended Actions 5
C.
Value-Impact Data on the Recommended Actions 1.
Risk Analysis Results 6
2.
Industry Impact 9
3.
NRC Operations 12 4.
Other Government Agencies 12 5.
Public 12 II. Regulatory Resolution A.
Regulatory Alternatives 13 B.
Discussion and Comparison of Regulatory Alternatives 13 III. Recommended Implementation Plan A.
Safety or Environmental Significance of Froposed Action 13 B.
Recommended Resolution Actions 13 IV. Statutory Considerations A,
NRC Authority 14 B.
Need for NEPA Statement 14 V.
Summary and Conclusions 14 References 16 Appendix A Public Comments Received and Action Taken A-1 v
REGULATORY ANALYSIS FOR USI A-1, WATER HAMMER I.
The Recomended Actions A.
Summary of Problem and Recommended Actions The Unresolved Safety Issue (USI) A-1 deals with safety concerns related to water hammer occurrence in nuclear power plants. The staff's concerns were prompted by the incteasing frequency of water hammer occurrence (see Figures 1 and 2) in the mid-1970's as new plants were coming on line, and, in particular, by the feedwater line rupture at Indian Point 2 in 1973 (attributed to water hammer induced by steam-void collapse).
Principal concerns were:
the potential for inadequate dynamic load design, disabling of safety systems, and the release of radioactivity.
The staff's views were set forth in NUREG-0582 (Ref. 1), and water hammer ias designated a USI in 1979.
Historically, nearly 150 water hammer events have been reported since 1969; 81 have occurred in boiling water reactors (BWRs) and 67 have occurred in pressurized water reactors (PWRs).
(Twenty-seven of the PWR water hamers have occurred in steam generators.)
With the exception of the Indian Point 2 event in 1973, reported damage has been principally confined to pipe hangers, snubber system::. and equipment-mounting structures.
Furthermore, approximately half of these water hamers occurred in the plant preoperational phase or first year of commercial operation (which indicates a plant operational learning process). Also, only about half of the operating plants have reported water hammer occurrences.
A compilation of reported water hammer occurrences, underlying causes and plant corrective actions taken is provided in NUREG/CR-2059 (Ref. 2).
As noted above, the increasing frequency of occurrence drew both staff and utility attention to water hammer, and corrective actions were implemented in the mid-1970's.
Steam generator (top feedring design) water hammer was studied (Ref. 3) and eliminated through NRC-initiated design retrofits calling for J-tubes, shortened piping, and controlling auxiliary feedwater flow rates (Ref. 4).
Design corrective actions were also initiated by the industry and implemented for BWRs (e.g., " keep-full" systems, vacuum breakers, etc.). The net result of the corrective plant design modifications has been a reduced frequency of water hammer occurrence.
I REGULATORY ANALYSIS Following first year of 39/81 events commercial operation 13/81 r$al peration
~
Co 29/81 Prior to comercial operation I
I f
I O
10 20 30 40 Number of Events 30 h
[
20 l
3 I
E 10 5
0-12 10 7
f O
/
/
/
/
69 70 71 72 73 74 75 76 77 78 79 80 81 82 Calendar Year Figure 1 Reported Water hamer occurrences in US BWRs.
REGULATORY ANALYSIS Non SG water hanner events Steam Generator WH events 23/40 events 15/27 events
,per on oper t on
~
First year of comercial 6/27 First year of comercial 6/40 operation operation Prior to comeretal Prior to comercial 6/27 11/40 operation operation I
I I
I I
I I
I O
10 20 30 0
1-0 20 30 Number of Events Number of Events 50 E
i 40 g
f I
, 30
%e
_0 20 a
1 10 I
O-7 Non SG
/
Water 10
/s Hamer r
E a
8 7l/
I 7/
t 6
a
/
E
/
4 F-/ f f F-
/
f k f f U
2 h S h VA
/
h
/
/
0 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 Calendar Year Figure 2
Reported water hamer occurrences in US PWRs.
- w.y..+;.,e p ++ q s
- m r
- ;. ;. & w s e t
- L.'
O.% '.
- . _ y p ;.-
't,
'} }.,,Y
.E.
y isj' L.; ;
+
y; REGULATORY ANALYSIS " ? *? ;
w.:
+..-x o
Q[, 'e.:,
u.:. e...
[
The water hammer issue has recently been further studied 'Ref. 5),
..i y
.1 7
and the technical conclusions derived reveal a significanlly lesser k.
safety concern than previously hypothesized.
These resul;s can be L 'i; 14 summarized as follows:
3 ;f ". '.
(1) Total elimination of water hammer is not feasible due to the
!..b ' ~
possible coexistence of steam, water, and voids in v'rious 1.6
^
subsystems. Experience shows that design inadequacies and
? h.# g operator-or maintenance-related actions have contributed F h }; ',
about equally to water hammer occurrences.
BWRs are i F i
intrinsically more susceptible to water hammer occurrence.
?-3 * ~
N.$. J (2) Reported damage has been principally confined to piping
- Q kg t
support systems, and none of the reported water hammer g... m 5-occurrences has resulted in any radioactive release.
. f Ni i.
(3) Frequency and severity of water hammer can be reduced and
',, z.3
'&p 3.f -
1 maintained low throuqh the continued use of the types of 3
design features discussed above.
h.'.N.s i, W i..
E (4) Additional operator awareness and training could lead to a
('"6M L
further reduction of water hammer occurrence.
Use of void h.rC.
~,s detection instrumentation to alert operators to voided 9
conditions would also help.
qfig. sN
(
M#p The staff's current technical findings relative to the water hammer b 5. 1 3) issue are set forth in NUREG-0927 (Ref. S).
These findings are
.I7' i
2.K f2 i
based on water hammer evaluations; References 2, 4, 5 and 8; and
}.,
public comments received (see Appendix A).
1 %:
7 The following actions are recommended:
m g.y 4
(1)
Issue the staff's water hammer technical findings (NUREG-0927) pg.S.4 as an informational document for use by the industry for
. g d.n feedback of design and cperating experience to plant staff.
Ed g
[
NUREG-0927 reviews water hammer occurrences, underlying
.c.O &
.. i causes, and systems affected, and sets forth potential means f.4 p.
for avoiding water hammer.
h"u, y
.a 3
(2) Ensure operator awareness and training (for avoiding water A
1%
hammer) through the implementation of TMI Task Action Plan,
.i k,' R Part I.C.5, " Procedures for Feedback of Operating Experience to Plant Staff," and Part I.A.2.3, " Administration of Training i
. N Prog rams. " The NRC Office of Inspection and Enforcement (IE)
' ~
has verified, through its inspection program, that general pN...
procedures for implementing Part I.C.5 have been established.
7 9 j '.'
The Licensee Qualifications Branch of the NRC Division of ifyi f.:
q
.,.P'..
~.'
Y e{.Y '
z
.w 4
- l
.p g~-),.,
O A
j ] ' ?.
-)
V ',
~.%,..
. > 1 p.. J - y.!. ; Q-Y:
- - 4
- ..Y W n:" g'.J:s ;(L ;.O:R 4.R. M.? :- M " ' i. : ' l-
REGULATORY ANALYSIS Human Factors Safety (DHFS) is developing guidelines and criteria to evaluate and upgrade utility training programs (per Part I.A.2.3) and will include water hammer as one of a number of safety issues currently identified. Since activities for implementing applicable sections of the TMI Task Action Plan are under way, and since the safety significance of water hammer is less than previously viewed, ro special action to implement findings presented in NUREG-0927 is necessary.
(3)
Issue the following revisions to Standard Review Plan (SRP)
Sections:
3.9.3, ASME Code Class 1, 2, and 3 Components, Component Supports, and Core Support Structures; 3,9.4, Control Rod Drive Systems; 5.4.6, Reactor Core Isolation Cooling System (BWR); 5.4.7, Residual Heat Removal (RHR)
System; 6.3 Emergency Core Cooling System; 9.2.1 Station Service Water System; 9.2.2, Reactor Auxiliary Cooling Water Systems; 10.3, Main Steam Supply Systems; and 10.4.7, Condensate and Feedwater Systems reflect current water hammer findings and will ensure continued use of design features which have eliminated or minimized water hammer occurrence.
Public comments received have been reflected in these SRP revisions (see Appendix A).
The revised SRPs would be used for reviews of " custom plant" Construction Permit (CP) applications and for reviews of Standard Plant applications docketed after issuance of the revision and which are intended for referencing in CP applications.
8.
Need for Recommended Actions The need for the recommended actions is as follows:
(1) Make use of experience gained regarding design features and operating experience which have shown a capability to eliminate or minimize water hammer occurrence to ensure that future plant designs utilize design features proved effective in eliminating water hammer.
(2) Clarify current staff review practices to ensure that the review process is more predictable and thus reduce the burden of the regulatory process.
Yh
I
?fQ ;
REGULATORY ANALYSIS '
'j
'%a.
C.
Value-Impact Data on the Recommended Actions
- ?.V d f
-) r R..
1.
Risk Analysis Results D.'h C.'
7 s.. '.: ' ;
A risk assessment study (Ref. 6) was performed to assess the W ~.[
significance of risk from water hammer occurrence with respect f 'C CI.! O to overall plant risk.
Water hammer frequencies were derived gg.
from reported occurrences (Ref. 2), and cnmponent or system
,M 4 failure models were developed from system assessments (Ref.
M 1' 5).
For exaniple, if only piping support damage was reported, 1 7Rj then the assumption was made that the system would still A;). -
function.
If water hammer occurrence resulted in disabling 7 9[7 the system, then models were constructed for modifying failure i
'g> i "
q event networks (i.e., failure-on-demand frequency).
A more c
detailed discussion of the derived frequencies and failure j
- N. -
models is contained in Reference 7.
4 i
p
- t.,.
Three specific. nuclear plants for which probabilistic risk
.]
assessment (PRA) models were available (namely Millstone 1 (BWR-3), Browns Ferry 2 (BWR-4), and Sequoyah 1 (PWR)) were E
kW{ 1 selected for this risk study as representative of operating reactors. Since reported water hammer experience reveals a 4 t higher frequency of occurrence in BWRs and a dependence on p~
Y@)i different BWR designs, the emphasis was directed at potential r
BWR risks.
1
'.: ~ :s
.c yn, Q Y )
?. '.
3 The release categories and associated public dose estimates
" tv employed are shown in Table 1.
These public dose values were
'n derived using the CRAC code and assuming the guidelines and quantities of radioactive isotopes used in the Reactor Safety I,,yy]%.f
- g. 7 Study (WASH-1400), the meteorology at a typical mid-west site (9%ti)
(Byron-Braidwood), a uniform population density of 340 people
.4 J..wg%
per square mile (which is an average of all U.S. nuclear power plant sites), and no evacuation of population. They are also Ziy i based on a 50-mile-release-radius model (see also Ref. 7).
'Kb" The release categories shown in Table 1 correspond to radiological release causes described in WASH-1400 (e.g.,
steam explosion with containment rupture, core melt, etc.).
The, estimated public dose due tr vater hammer was derived from the " base case" PRA results versus calculated increases in core melt frequencies and increases in the respective release category frequencies for the plants noted above (see Table 1).
Basically, the calculations provide a means to compare calculated risk results with and without water hanner.
s REGULATORY ANALYSIS b eis
[. b a.,
%, 4 The results of these risk assessments are summarized in Table 2, where both calculated public doses and core melt Qf/i' frequencies are shown. The differences shown'in the third 7d%
column are the calculated change due to inclusion of water i* di f.[Q-hammer-induced failures in the event trees.
These calculations are conservative since the assumption was made that safety systems were disabled as a result of a frequency-
%r of-failure or demand model as derived from reported water
- ?.).
hammer events.
(Refer to Reference 6 for a more detailed
'N 4 -
analysis.)
The results in Table 2 can be summarized as follows:
(1) Water hammer effects on PWR risk are negligible.
(2) Water hammer effects on BWR risk are negligible or small.
As part of the risk analyses performed for BWRs, BWR plants with isolation condensers (ICs) were evaluated in some detail because:
(a) Millstone 1 has repeatedly incurred water hammer in the IC, and (b) should a water hammer fail the pressure boundary of the IC, a direct release pathway is opened to the environplent. This type of failure model in a risk analysis using the Millstone Integrated Reliability Evaluation Program model resulted in a significant dose and consequent risk from potential IC failure by water hammer.
When the risk model was modified to include a feedwater pump trip on high reactor vessel water level, the risk from water hammer in the isolation condenser was virtually eliminated.
The risk analyses, therefore, showed that a high reactor vessel water level feedwater pump trip, which removes the conditions for water carryover into the IC, is a generic resolution to the problem.
Operating experience data support this conclusion.
Plants that have a feedwater pump trip (Dresden 2 and 3) have not reported water hammer in the IC. Some plants without such a trip (Millstone 1 and Nine Mile Point) nave reported IC water hammer events. Millstone 1 has not reported an IC water hammer since installation of the feedwater pump trip about 10 months ago.
Table 3 provides an overview of all operating BWR plants with ICs. Only Oyster Creek and Big Rock Point have not installed or have not committed to have installed a high reactor vessel water level feedwater pump trip.
Neither plant has reported any water hammer experience with its IC. As noted in the
}:: ' i
.. g
.,%, u REGULATORY ANALYSIS ' in.
- w -
-a ::,yn,.
@S.~-
TABLE 1, PUBLIC DOSE VALUES UTILIZED FOR USI A-1 RISK STUDY
'.Vf Release Release Category Dose YH Category Multipliera(man-rem)
?.lif PWR-1 5.4E+6
.4 > "., J l.r.j:t PWR-2 4.8E+6 f
PWR-3 5.4E+6
F DF.
%@ P PWR-4 2.7E+6
'T PWR-5 1.0E+6 PWR-6 1.5E+4 fjR(i PWR-7 2.3E+4 s.J;.)
PWR-8 7.5E+4
- 33. S.
PWR-9 1.2E+2
-dA.(
BWR-1 5.4E+6
.W AE:(.,...J BWR-2 7.1E+6
??
BWR-3 5.1E+6 BWR-4 6.1E+5 BWR-5 2.0E+1
- Values from NUREG/CR-2800 (7); (man-rem) x (probability of occurrence) = public dose resulting from the release category noted. Total release obtained by summing the categories.
TABLE 2
SUMMARY
OF RISK ASSESSMENT CALCULATIONS Calculated Public Dose (man-rem / plant-year)
Type of Base Case With Water Calculated Increase Plant (w/o W.H.)
Hammer (due to W.H.)
BWR-3s No calculated change due to Water Hammer.
BWR-4s 1147 1169 22 PWRs No calculated change due to Water Hammer.
Calculated Core Melt Frequency (1/ plant-yrs)
Change in Core Type of Base Case W/ Water Melt Frequency Plant (w/o W.H.)
Hammer (due to W.H.)
BWR-2s No calculated change due to Water Hammer.
BWR-4s 2.0E-4 2.1E-4 1.0E-5 PWRs No calculated change due to Water Hammer.
em
.C
I REGULATORY ANALYSIS footnote to the table, Oyster Creek is expected to install such a trip for other reasons.
Big Rock Point is an older, smaller plant whose overall safety design is being addressed in the Systematic Evaluation Program.
Therefore, no special or additional action for BWRs with ICs is contemplated as a part of the resolution for A-1.
It should be clearly recognized that the dose and risk attributable to IC water hammer are calculated values.
None of the reported water hammer events has resulted in a significant release of radioactive material to the environment.
Using the incremental dose due to water hammer shown in Table 2, and assuming an " average" outstanding plant life of 25 years, the following change in public dose can be calculated:
BWR-4 Averted Public Dose
= (22 man-rem /Rx-yr) (25 yrs)
= 550 man-rem /Rx BWR-3s = No calculated change due to water hammer PWRs = No calculated change due to water hammer.
These increases in public dose can be viewed as " averted public dose" (presupposing that corrective action is taken to avoid water hammer) for value-impact discussions. Thus, the very low values (0-550 man-rem / reactor) calculated for both PWRs and BWRs, do not support any special hardware backfit actions for operating plants.
2.
Industry Impact No new plant hardware or design changes are being recommended as a result of the USI A-1 resolution evaluations.
The feedwater pump trip (noted previously as providing a generic resolution to BWR isolation condenser water hammers) is either in place or is being installed for other reasons in BWR-3s with ICs. Therefore, plant impacts are judged to be minimal or nonexistent.
TMI Task Action Plan I.C.5, " Procedures for Feedback of Operating Experience to Plant Staff," requires that procedures be developed for feedback of operating experience to plant staff.
Several groups within the industry (e.g., Institute of Nuclear Power Operations and reactor owners groups) have taken the lead in providing collective competance for meeting I.C.5 requirements.
IE has verified the establishment of general g
guidelines for implementing I.C.5.
Issuance of NUREG-0927 for
}
w REGULATORY ANALYSIS TABLE 3 BWRs with Isolation Condensers Number of Independent Installed Pump Trip Isolation On High Vessel Plant Type Condenser Loops Water Level Millstone 1 3WR-3 1
Yes Dresden 2 BWR-3 1
Yes Dresden 3 BWR-3 1
Yes Oyster Creek BWR-2 2
No
- Nine Mile Point 1 BWR-2 2
No*
Big Rock Point BWR-1 2
No
- These plants will in all likelihood need to install vessel overfill protection to reference Generic GE Safety /Relicf Valve testing in their responses to TMI Action Plan Part II.D.1.
Nine Mile Point is now committeJ to installing the trip in 1984.
1 T
REGULATORY ANALYSIS MY
- {dt informational purposes will assist industry activities h
currently under way.
A more comprehensive set of guidelines yo and criteria to evaluate and upgrade utility training programs (per TMI Task Action Plan, I.A.2.3) is being developed by the Q.C ~
Licensee Qualifications Branch. Water hammer is one of a
$i number of safety issues being identified in the Licensee
, g. f,
Qualification Branch plan. Thus issuance of NUREG-0927 will s,'
provide information which can be used with both I.C.5 and J, ' ' "
I.A.2.3.
- ?.
With respect to forward fits (f.e., implementation of the SE.1 revised SRP Sections), the impact should also be minimal. The p$
proposed changes reflect design changes which have come about 4....
to remedy specific water hammer occurrences (i.e., fixes for Jcy top feedring steam generators, etc,) as problems arose and
. (J.:
therefor: represent proven design concepts. Since the h
proposed SRP revisions reflect the current state-of-knowledge r ::}
%j%)c concerning water hammer occurrences and systems which can d
prevent or minimize water hammer, the designer / operator can incorporate these revisions of proven system design changes.
jf; wp~
In addition, the SRP has not previously contained specific guidance for reviewers with respect to water hammer considerations, with the exception of review guidance for water hammer in top feedring steam generators (ASB BTP 10-2).
Thus the depth and scope of staff review have varied with individual reviewer experience and insights; however, this is consistent with the audit nature of the staff's review function. These changes do identify water hammer review areas that should be addressed on the basis of prior water hammer occurrences, design changes implemented by industry, and precautionary measures indicated by operating experience.
Thus revising these SRP sections to include specific guidance on water hammer will clarify staff review practices and ultimately reduce the burden of the regulatory process.
NUREG-0927 (which summarizes findings based on water hammer experience) can be used as a reference technical report.
Thus industry impact is judged to be minimal.
Design costs associated with avoiding water hammer could be on the order of
$50,000-100,000 (0.5-1.0 year of engineering time).
The cost of new systems such as keep-full systems ($200,000-400,000),
vacuum breakers ($100,000-300,000), and feedwater control systems ($100,000-200,000) are not insignificant, but they do not constitute major plant equipment costs.
Operator training for water hammer avoidance is estimated to be on the order of
$25,000-50,000 per plant.
(The preceding cost estimates are
5 k
REGULATORY ANALYSIS e b
h based on discussions with vendors in the nuclear industry and should be viewed as preliminary estimates.
Firm cost 1
estimates will require plant-specific design take-offs for T
i estimating actual equipment and associated installation costs.)
k 3.
NRC Operations h
The impact on NRC operations (or the review process) is I-negligible. The proposed SRP changes will reflect licensing C
review positions for new construction permit applications, y
since only a " forward fit" is recommended. NUREG-0927, the 2
technit.al findings report, will be of use to both the reviewer 2_
and applicant. The estimated impact for reviewer attention to water hammer using the proposed SRP revisions is 4 weeks of 6"-
engineering effort (or $15,000/ plant).
5 Y
With respect to followup to TMI Task Action Plan, Part I.C.5, a
each utility is required to conduct an internal audit to 2
ensure that the feedback program functions at all levels. The 2
Licensee Qualifications Branch is currently developing 1
guidelines and criteria to evaluate and upgrade utility 52 training programs to implement TMI Task Action Plan, Part
'i I.A.2.3.
Water hammer (as well as other safety issues) will 4.
be included in these guidelines. The Offices of Nuclear j-Raactor Regulation and Inspection and Enforcement will monitor 4
effectiveness of this approach following iniplementation.
(5 4.
Other Government Agencies K
y No impact on other government agencies is projected.
b 5.
Public The value to the public would be the avoidance of public dose 3
associated with water hammer events leading to core melt and
?
offsite releases.
As discussed in Section I.C.1 (see also f
Table 2), the calculated additional releases are zero for PWRs i
and approximately 550 man-rem / plant for BWRs.
Since averted y
dose estimates are negligible, the principal value to the public would be to provide feedback to the industry in experience gained and in maintaining proven design concepts for future plants.
L-;
i M
r r
N
~
REGULATORY ANALYSIS II.
Regulatory Resolution A.
Regulatory Alternatives (1)
Issue NUREG-0927 for information only, and follow up on the feedback of operational experience through implementation of the TMI Task Action Plan, Part I.C.5, (2)
Issue NUREG-0927 for information, and issue for comment proposed revisions to SRP Sections 5.4.6, 5.4.7, 6.3, 9.2.1, 9.2.2, 10.3, and 10.4.7.
These revisions reflect design changes and operating procedures which have proven effective in preventing water hammer. The revised SRP sections would apply to new custom plant CP reviews.
B.
Discussion and Comparison of Regulatory Alternatives (1)
Issuance of NUREG-0927 for information only would have a zero impact.
Followup on incorporation of wate. hammer operating experience into training models by the NRC Regional staff, per TMI Task Action Plan, Part I.C.5 would be a minimal impact since such inspections are normally done.
(2) Option (2), revision of the SRP sections noted and issuance of NUREG-0927 (for information) is a minimum impact and a forward fit approach. This option would ensure that future CP reviews will consider water hammer findings and design features currently proven effective to avoid water hammer. This is the recommended course of action.
III. Recommended Implementation Plan A.
Safety or Environmental Significance of Proposed Actions The principal safety significance rests primarily on continuance of established plant design and operational procedures that have demonstrated the capability to minimize or avoid water hammer severity and damage, thereby avoiding damage leading to radioactive release. The recommended approach (Option 2) provides for continuance of proven design and operational considerations.
B.
Recommended Resolution Action (1)
Issue NUREG-0927 containing the staff's technical findings as a technical information document.
t-1-
REGULATORY ANALYSIS -=
-i (2)
Issue and implement the revised SRP Sections 3.9.3, ASME Code 7
Class 1, 2, and 3, Components Supports and Core Support
=-
Structures; 3.9.4, Control Rod Drive Systems; 5.4.6, Reactor Core Isolation Cooling System (BWR); 5.4.7, Residual Heat L
Removal (RHR) System; 6.3, Emergency Core Cooling System; 9.2.1, Station Service Water System; 9.2.2 Reactor Auxiliary Cooling Water Systems; 10.3, Main Steam Supply Systems; and T
10.4.7, Condensate and Feedwater Systems, which are based on 3
the findings reported in NUREG-0927, public comments received, y
and concluding staff evaluations.
Implementation of these revised SRP sections will apply to the h
review of custom plant CP applications and standard plant
+
applications that are docketed after issuance of the SRP c--
revisions.
These revised SRP sections incorporate licensing review guidance which would maintain use of experience gained 7
regarding plant design features proved effective in eliminating water hamer, and which also clarify current staff i
review practices to ensure that the review process is more E
predictable and definitive.
(3)
Issue NUREG-0993 (formerly issued for public comment as the E
value-impact analysis) as the Regulatory Analysis in support E
of the SRP revisions noted above.
g IV.
Statutory Considerations h&
A.
NRC Authority D
e-The recommended changes to SRP Sections 3.9.3, 3.9.4, 5.4.6, 5.4.7v
?
6.3, 9.2.1, 9.2.2, 10.3, 10.4.7 are within the statutory authority E
of the NRC. Also, plant-specific safety assessments are within the statutory authority of the NRC.
7 B.
Need for National Envirnnmental Policy Act (NEPA) Statement g
The proposed changes do not warrant a NEPA stai.ement.
O V.
Summary and Conclusions Based on the above discussion, the following actions are recommended:
7--
4 (1)
Issue the revised SRP Sections for forward-fit implementation.
5
=m 1--
REGULATORY ANALYSIS (2)
Issue NUREG-0927 as a technical findings document.
This staff report summarizes the staff's assessment of water hammer in nuclear power plants.
(3) Ensure operator awareness and training with respect to avoiding water hammer through the use of the TMI Task Action Plan, Part I.C.5 and Part I. A.2.3, oprator training evaluatior. criteria under current development by the Licensee Qualifications Branch.
(4) Conclude current Operating License reviews through staff evaluations in progress.
REGULATORY ANALYSIS REFERENCES (1)
U. S. Nuclear Regulatory Commission, NUREG-0582, " Water Hammer in Nuclear Power Plants," April 1979.
(2) Chapman, R. L. et al., " Compilation of Data Concerning Known and Suspected Water Hammer Events in Nuclear Power Plants," report prepared by EG&G Inc. for the NRC, NUREG/CR-2059, May 1982.
(3) Block, J. A. et al., "An Evaluation of PWR Steam Generator Water Hammer," report prepared by Creare Inc. for the NRC, NUREG-0291, June 1977.
(4) Anderson, N. and J. T. Han, Prevention and Mitigation of Steam Generator l
Water Hammer Events in PWR Plants NUREG-0918, November 1982.
l (5) Offer, R. A. et al., " Evaluation of Water Hammer Events in Light Water l
Reactor Plants," report prepared by Quadrex Corp. for the NRC, NUREG/CR-2781, July 1982.
(6) Amico, P. and W. Ferrell, "Probabilistic Assessment of Unresolved Safety Issue A-1: Water Hammer," report prepared by Science Applications, Inc.,
September 1982.
(7) Andrews, W. B. et al., " Guidelines for Nuclear Power Plant Safety Issue Prioritization Information Development," NUREG/CR-2800, February 1983.
(8) Uffer, R. A. et al., " Evaluation of Water Hammer Potential in Preheat Steam Generators," report prepared by Quadrex Corp. for the NRC, NUREG/CR-3090, December 1982.
(9) Serkiz, A.
W., " Evaluation of Water Hammer Occurrence in Nuclear Power Plants-Technical Findings Relevant to Unresolved Safety Issue A-1,"
NUREG-0927, Rev. 1.
March 1984.
l l
APPENDIX A A-1 l
I APPENDIX A
SUMMARY
OF PUBLIC COMMENTS RECEIVED AND ACTION TAKEN T
Appendix A Public Comments Received and Action Taken Coment NRC Staff Response Stone & Webster Engineering Corp. (S&W) (7/15/83):
The practice of including classical water The principal thrust of the S&W recomendations to l
hamers as occasional mechanical loads (per ASME maintain the practice of designing for water hamer B&PV Code,Section III) in piping stress analyses loads in piping stress analysis and piping supports and piping system design support requirements has been incorporated into NUREG-0927, and SRP 3.9.3, should be maintained.
"ASME Code Class 1, 2, and 3 Components, Component Supports, and Core Support Structures," has been revised to reflect a need to consider water hammer Include a statement to this effect in the revised in development of Design Specifications under the
{
SRP sections, which had been issued for coment.
ASME Code requirements.
Issue NUREG-0927 for infonnation, along with the SRP revisions proposed.
Westinghouse Electric Corp. (W) (7/20/83):
I Westinghouse recommended that the concept of design Westinghouse's points regarding verification of design qualification of qualified designs within specified adequacy relative to water hamer qualification testing operating limits without requiring in-plant testing (lab and first plant) as the basis for acceptance of for each application be reflected in NUREG-0927 repeated applications of the same design in subsequent and the SRP. Westinghouse goes on to make the point plants is technically scund; it also is representative that once a design has been qualified within of the means by which new designs are introduced into specified operating parameter limits to preclude SAR submittals. Although the staff agrees that qualified or minimize water hamer, adequate assurance exists standard designs for steam generators may be available, that repeated applications of the same design the main feedwater and auxiliary feedwater systems and within the same limits will have minimum potential their controls are, in general, plant-specific and for water hamer.
their design repeatability has been limited to identical plants at one site, or to standardized plants, if these systems are all within the scope of the Preliminary Design Approval.
For these cases, it has been the practice of the NRC staff to not require that the testing in BTP ASB 10-2 be repeated on the identical plant. However, in general, the steam generator main and auxiliary feedwater system and controls and their operating procedures are s
h s
Appendix A Continued Comment NRC Staff Response are not the same and even modest changes have resulted in steam generator water hamer (such as the Maine Yankee water hamer occurrence in 1983).
The stiff has concluded, therefore, that it is prudent to maintain BTP ASB 10-2 to demonstrate for new plants the absence of a proclivity to water hamer by preoperational testing.
W also comented on the feedwater control valve The W coments regarding verification that the relative to feedwater line water hamer in PWRs feedwater control valve is compatable with other and recomended that compatability of feed system systems and control requirements have been components should be verified by the system incorporated in NUREG-0927.
designer.
W made several editorial coments (see W coments ilos. 3,-4, 5, 6, and 7 in Enclosure 3). _
NUREG-0927 has been modified as appropriate.
Reactor Controls, Inc. (RCI) (7/29/83):
RCI addressed concerns regarding water hammer load RCI's recommendation to address potential water during the scram function of the control rod hamer in BWR CRD systems has been incorporated drive (CRD) hydraulic system.
RCI cited testing into SRP 3.9.4, " Control Rod Drive Systems,"
testing and analysis of the CRD system that has I&E will handle matters related to CRD water shown that water hammer loads require hammer in operating plants, and NRR will continue consideration.
RCI recomended that either SRP licensing review activities (per TIA 82-59).
3.9.4 or SRP 4.6 be revised to include specifically the need to consider adverse dynamic loads such as water hamer in the CRD system.
I i
hmgMMEMC&BM
Appendix A Continued Coment NRC Staff Response J. F. Doherty (8/7/83):
Mr. Doherty commented that NUREG-0927 should address Mr. Doherty's concerns related to BWR pipe cracks the likelihood of pipe break on certain pipes and pipe failures are of a broader nature than USI (particularly in the BWR) with given cracks, instead A-1 and are being addressed by current regulatory of assuming completely intact pipe and assessing reviews related to reactor piping.
NRC's Piping Review d,amage to pipe supports and restraints.
Comittee is currently reviewing regulatory practices related to pipe cracks, pipe breaks, seismic design, and dynamic load / load combinations. Water hamer loads are one of several dynamic loads that will be reviewed with respect to experience and current design practices. Any recommendations for changes in regulatory requirements for the la design of piping and piping supports will include consideration of their effect on the capability of piping systems to withstand water hammer loads.
Fluid Components, Inc. (FCI) (8/9/83):
l l
FCI raised questions regarding water hamer FCI's comments regarding water hamer in degraded in BWR piping that has incurred intergranular BWR piping are similar to Mr. Doherty's.
BWR stress corrosion.
pipe cracks and pipe failures are undergoing a major regulatory review from a more comprehensive FCI also pointed out problems in the use of safety viewpoint.
See also the response above.
level sensors, as shown in Figure 2.4 and 2.5, and I
recomends the use of a thermal dispersion level j
switch. The thermal dispersion principle also can warn the operator of incipient voiding.
Review and Synthesis Assoc., S. H. Bush (11/1/83):
Mr. Bush commented:
"In my opinion, the Staff The staff has neither assumed that previous water position with regard to water hamer is hamers represent upper bound energy levels, nor unrealistically optimistic.
It works on the a has the staff based its concluding evaluations on priori assumption that previous water hamers in calculated energy loads.
Rather, the probabilistic nuclear systems represent bound energy levels. This risk assessment (PRA) performed for USI A-1
Appendix A Continued Comment NRC Staff Response is based on after-the-fact calculations of energy utilized a failure model developed from reported levels for a limited number of water hammer or water water hammer occurrences (or frequency) and water slugging events. These calculation yielded values hammer occurrences (or frequency) and that are a small fraction of the theoretical energy underlying causes. The PRA conservatively assumed bound. While I do not anticipate cases of water that a predicted limiting event (such as pipe hammer near the theoretical upper bound, I would not rupture or loss of safety system function) would be surprised if some were to accur that are several occur even for those systems in which no disabling times the existing calculated levels."
event was reported. The calculated change in public dose due to water hammer induced failures was negligible or very low (i.e., less than 25 ma n-rem /rx-yr).
?>
In addition, it should be noted that design basis accidents (DBAs) currently assume double ended en guillotine breaks (DEGBs) for analyzing accident scenarios and the design of safety systems.
Regardless of the postulated upper bound loads, pipe breaks more severe than a DEGB are not likely to occur.
Mr. Bush also stated:
"The Staff also labors under Water hammer has not been dismissed on the basis of some misconceptions that aren't necessarily valid.
redundancy. The staff has carefully reviewed water There is a concern for the global effect of seismic hammer occurrences, their underlying causes, and events while dynamic events such as water hammer corrective actions taken, and finds that certain are dismissed on the basis of redundancy, etc.
In design fixes (e.g., J-tubes in top feedring steam the following, I shall attempt to cite some positive generators, keep-full systems and vacuum breakers aspects, followed with points where I find the Staff in BWR systems) and operational awareness to avoid position unrealistically optimistic.
I shall use an conditions leading to water hammer have significantly example, not necessarily valid that attacks their reduced the frequency of water hammer. The staff's rather cavalier disnissal of such events....
findings are reported in NUREG-0927.
Also, there is an important difference between global events (such as earthquakes) and water hammer events. An earthquake event can affect many lines and systems simultaneously. The effect of a water hammer occurrence is generally limited to a single system and its attachments.
h
i Appendix A Continu::d Coment NRC Staff Response
" Events such as the H. B. Robinson and Turkey Point Inadequate designs are prone to failure, failures where dynamic loacs resulting from relief Corrective measures have been taken and proved to valve closure blew the valves off the header were be effective.
corrected by modifying an admittedly lousy weld 1
joint design. Apparently, the industry learned a i
lesson because we haven't had any more such failures.
"In the early 1970's, there were a large number of Keep-full systems have essentially eliminated water water slugging events where a valve was opened hammer occurrence in BWRs. Before jockey pumps into a voided line and pipes were bent, hangers were used, line voiding was a major cause of water 7
pulled out of the wall, etc. Techniques such as hammer in BWRs. The staff supported industry's.
jockey pumps to fill the voided lines have markedly introduction of such systems and has recomended -
minimized such events.
I'm doubtful they have continued use of this design feature. The staff totally eliminated them.
has stated on numerous occasions that total elimination of water hammer is not possible.
" Steam-water reactions, particularly those induced J-tubes have been shown effective in minimizing from the steam generator, have been virtually steam generator water hammer (SGWH) in some PWRs.
eliminated through installation of J-tubes. Again, It should be noted that not all PWRs have installed I am not sure that steam-water reactions have been J-tubes and that some plants-have operated without eliminated."
J-tubes since comencement of commercial operation without water hammer occurrence. NRC's plant review actions and conditional approvals for continued plant operation are summarized in NUREG-0918. Maine-Yankee was one such plant granted approval to operate.
Following the January 1983 water hammer occurrence at Maine Yankee, J-tubes have been installed, operating procedures revised as needed, and a pre-operational test run to demonstrate absence of water hammer under loss of feedwater conditions. Again, total elimination of SGWH is not claimed.
+
'k ";.l kke M; sy r : K y
. k.i 4 ; ; ).; w. i b u '
v,s.) :L, y Appendix A Continued Cor:rnent NRC Staff Response Mr. Bush went on to state:
"As indicated, I am The staff is aware of deaerator systems in fossil doubtful we have experienced feasiLie, more fuel plants that have experienced numerous large energetic water hammers. An examination of the water harrrners. These events have not resulted in industrial literature will reveal water hamers catastrophic piping failures even though severe that catastrophically failed piping. These are damage has occurred in piping support systems and still possible."
building structures.
In these systems, piping cracks have occurred near piping anchors as a result of frequent large bending moments introduced ty long segments of flexible piping systems.
However, nuclear power plants are built' to more
!:n stringent codes and standards than are nomally employed in many industrial piping systems. The water harrrner data base for USI A-1 represents more than 600 reactor-years of operating experience, over which time fewer than 150 water hamers have occurred.
Since 1982 only two water hammers have occurred; one of these occurred in pre-operational testing. With the exceptions of the IP-2 feedwater line rupture in 1972 and the Maine Yankee feedwater nozzle crack in 1983, all other water hamer damage has been primarily to piping and equipment supports.
Catastrophic water hammers have not occurred in nuclear power plants.
" Earlier incidents of water hamer or water slugging Water hamer occurrences were not dismissed because tended to be dismissed they 'only' pulled out all they only pulled out all supports for a few hundred supports for a hundred feet or more of piping feet...," or for any other reason. The staff's concerns rather than failing the pipe. These support regarding water hammer occurrences in the early and failures served as excellent energy absorbers mid 1970s are well documented in NUREG-0582 (July 1979) minimizing damage to piping. Since then, we have and resulted in water hammer being designated in unresolved gone in the wrong direction:
namely, using large safety issue (USI A-1).
As noted above, design changes embedment plates, larger bolts, bigger lugs on the and operation awareness of water hammer potential piping, etc. These measures almost certainly have significantly reduced occurrence frequency transfer the energy absorption to the pipe.
In (see also NUREG-0927).
I I'
l I
I I I I
I e
Appendix A Continued Comment NRC Staff Response the opinion of the PVRC Steering Committee and some of.the prestigious consultants for the NRC Task Group on Seismic Design, ASME III has gone in the wrong direction. The piping supports are too strong and the equipment supports too weak.
Bosnak's group feels the same. Hopefully, we can change it in time, but that probably will be for new plants, not a backfit requirenent. We may wish to make it a requirement along with requiring removal of excessive supports based on increased damping values.
"Another problem pertains-to the BWR IGSCC in larger Postulated changes in ASME codes should be addressed
'.u pipes. The new appendix to ASME XI addresses the by the appropriate ASME Committee, not USI A-1.
NRC seismic case.
I'm not sure we will see the same staff members and consultants are active members margins for more severe dynamic events.
Ev of such Committees.
Rodabaugh raised this concern and I intend to take it to ASME XI for consideration. There are other concerns we need to address with regard to this appendix. We probably acted too precipitiously,
- but there was a very real need.
"Let me address another concern.
The Staff The staff does not dismiss water hammer on the dismisses water hammer on the basis of redundant basis of redundant systems (see NUREG-0927).
systems.
Let me postulate a relatively unlikely, Moreover, the steam generator scenarior presented but not impossible, scenario.
It we hed a steam-by_ Mr. Bush is not possible. The occurrence of water reaction'at the feedwater-steam generator a steam-slugging water hammer in the feedwater interface, we could get a shock wave traversing line requires that the feedwater sparger be the pipe. We may get one in the steam generator uncovered. -Under such circumstances, any pressure that could break several tubes.
This is a classic wave emanating from the feedwater line would initiator for pressurized thermal shock and all be reduced to an insigificant magnitude upon the redundant systems available won't help in a entering the steam medium.
Even if the sparger
Appendix A Continutd Comnent NRC Staff Response reactor pressure vessel with a high transition were not uncovered, the pressure wave would be i
temperature.
While I admit this scenario is greatly reduced by the orifice effect of the sparger i
.unlikely, it points out the weakness in the staff holes or J-tubes, the extreme area difference between position.
the. steam generator and the feedwater line, and.the air chamber effect of the steam blanket over the water.
The effects on the steam generator tubes should be of the same order of magnitude or less than those resulting from the bubble collapse phenomenon, that occurs in steam generators following load rejection or turbine trip.
These events occur far more frequently than steam generator water hammer and do not fail tubes.
"With regard to the ACRS questions, I suspect a A PRA analysis dealing with the safety significance meaningful PRA would be extremely expensive and of water hammer has been perfonned (see SAI's report:
much more difficult than the LLNL PRA's on pipe "Probabilistic Assessment of Unresolved Safety Issue failure.
Inputs would be virtually non-existent A-1, Water Hanner," January 1983), and the results with the exception of events such as turbine trip have been sunnarized in NUREG-0993. As noted above, and valve closure.
Furthermore, the upper bound evaluation of water hammer events resulted in a values would be virtually impossible to live with, negligible change in calculated. risk.
" Summarizing,' water hammer problems have been The staff's position as reported in NUREG-0927 and reduced but not eliminated; the Staff position NUREG-0993 is not unduly optim.istic; the staff strikes me as unduly optimistic; positive action makes no claim to total elimination of water hammer.
may be necessary to correct the multiple problem The staff does support continued use of proven l
of too many supports and too strong supports.
design concepts and operational procedures that have significantly reduced water hanner occurrence, and i
such water hammer considerations are reflected in the revised sections of the Standard Review Plan.
In addition, an'NRC Piping Review Committee is currently reviewing regulatory practices related to pipe cracks, pipe breaks,. seismic design, and dynamic load / load combinations. Water hanner loads are one of several dynamic loads that will be reviewed with respect to experience and current
Appendix A Continued Comment NRC Staff Response design practices.
Thus, any recommendations for changes in regulatory requirements for the design of piping and piping supports will include consideration of their effect on the capability of l
piping systems to withstand water hammer loads.
E. C. Rodabaugh Associates Inc. (11/10/83):
l l
Mr. Rodabaugh-referred to discussions with S. Bush Mr. Rodabaugh's concerns related to stiffer piping
.and J. O'Brien and offers his opinion that water restraints and ASME Code Section XI, IWB-3640, will hammer is more of a concern than earthquake.
be considered by NRC's Piping Review Committee.
His Mr. Rodabaugh stated also that piping in newer views related to water hammer concerns outweighing.
plants is more restrained than in older plants earthquake concerns derive from to his participation e
that this may be of concern where the restraint-pipe in structural analysis working groups (in the mid-attachment involves lugs welded to the pipe pressure West), while similar opinions have been expressed.
boundary, should water hammer occur.
He recommends Although he has recommended additional research, he additional research to better define the water also states that great urgency does not exist for hammer problem to (1) identify potential water water hammer research.
NRC-RES will review his hammers too seJeare to " design against"; (2) identify recommendations.
the role of p;*sticity in evaluation of pressure boundary failure; and (3) determine if water hammer tests could feasibly be added to dynamic-loading-of-piping programs.
'0""
URYG0993 '$
1 U.S. NUCLEAD REGULATORY COMMISSION BIBLIOGRAPHIC DATA SHEET 4.TlhE AND SU8 TITLE (AddVolu Na, if mororissel
- 2. (Leave b/mkl 1
Regulatory Analysis for SI A-1, " Water Hamer"
- 3. RECIPIENT CESSION NO.
- 7. AUTHORIS)
- 6. DATE R[ ORT COMPLETED
"% ember IM A. W. Serk12
- 9. PERFORMING ORGANIZATION NAME AN AILING ADDRESS (/nclude tip Code)
DM REPORT ISSUED l
Division of Safety Technology
- m.
lI Office of Nuclear Reactor Reau ion
/ March t
U.S. Nuclear Regulatory Commiss' gg,,,, y,,,,,
Washington, DC 20555 f
f e. aem W=&s
^
'd) Tis" ion"'" f S#n'y^'fe"ch"noTog"y "^' '"
- 10. PROJECT / TASK / WORK U IT do.
TACS #04469 Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission
- 11. FIN NO.
Washington, DC 20555
- 13. TYPE OF REPORT P IOD COVE RED (/nclusive danes)
Final
- 15. SUPPLEMENTARY NOTES
- 14. (Leave Wmk/
- 16. AISTR ACT (200 words or less)
NUREG-0993, Revision 1 is the staff's regul. ory ana is dealing with the resolution of the Unresolved Safety Issue A-1, Water Hamer.
- This repor contains the value-impact analysis for this issue, public comments received,. nd staff res
.nse, or action taken, in response to those comments. The staff's technical f ' dings regardin. water hammer in nuclear power plants are contained in NUREG-0927.
/
- 17. KEY WORDS AND DOCUMENT ANALYSIS '
17a. DESCRIPTORS USI A-1 Water Hammer Regulatory Analysis for USI k-1 17b. IDENTIFIERS /OPEN-ENDED T S
- 18. AVAILABILITY STATEMENT 19.
CU n,s reportl
- 21. NO. OF PAGES Unlimited
- 20. SECURITY CLASS (Tass pp)
- 22. PRICE Unclassifind S
N RC FOXM 335 lit ett
7-UNITE'J STATES CUCLEAR REIULATORY COMMISSION PosNsYb$s5No WASHINGTON, D.C. 20565
,vsanh
,s n.... m 0FFICIAL SUSINESS PENALTY FOR PRIVATE USE,4300 1
I I
t 1ANIA119XLLSI i
120555078877
{
SOM VO l
T BR-PDR NOREG E PU I
(0 jY DC 20555 WASHINGTON 4
4 5
i l
I b't h.